Oriented Pattern Analysis for Streak Detection in Dermoscopy Images

Sadeghi, Maryam; Lee, Tim K.; McLean, David I.; Lui, Harvey; Atkins, M. Stella

doi:10.1007/978-3-642-33415-3_37

Cited by 13 publications

(17 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method for recognizing partial and complete radial streaming patterns builds upon and extends the earlier work [14], [15], through identi cation of valid streak lines from the set of candidate streak lines, to reduce false positive streaks such as hairs and skin lines. The method also extends the analysis to identify the orientation and spatial arrangement of streak lines.…”

Section: Methods For Three-class Classification For Streaks Detecmentioning

confidence: 98%

“…Our previous method for detecting the presence and absence of streaks [14], [15] was tested on a large database of 300 skin lesions. Because this paper's algorithm to classify a lesion into Absent, Regular, and Irregular is an extension of our previous work [14], the previous algorithm which utilizes a ngerprint recognition model is brie y outlined here.…”

Section: Previous Workmentioning

confidence: 99%

“…Because this paper's algorithm to classify a lesion into Absent, Regular, and Irregular is an extension of our previous work [14], the previous algorithm which utilizes a ngerprint recognition model is brie y outlined here.…”

Section: Previous Workmentioning

confidence: 99%

“…These valid streaks are ordered in the direction of the red arrow from 0 to for feature extraction in the next step. Based on the mathematical de nitions of streaks proposed in [14], a new set of 18 features is proposed for streaks, called STR (streaks), which includes three Structural, three Geometric, six Orientation, and six Chromatic characteristics of valid streaks. Common color and texture features [35], [18] of the lesion itself, called LCT (lesion color texture) has also been used.…”

Section: A Identifying Valid Streak Lines From Candidate Streaksmentioning

confidence: 99%

“…Such a system would not only help physicians diagnose melanoma as a second reader, it could also teach the technique of dermoscopy. This paper presents a novel automated method, which builds upon and extends the earlier work [14], [15] identifying the presence of valid streak lines from the set of candidate streak lines. This method analyzes further the orientation and spatial arrangement of streak lines and classi es the lesion as a lesion with Absent, Regular, and Irregular streaks.…”

mentioning

confidence: 93%

See 4 more Smart Citations

Detection and Analysis of Irregular Streaks in Dermoscopic Images of Skin Lesions

Sadeghi

Lee

McLean

et al. 2013

IEEE Trans. Med. Imaging

112

View full text Add to dashboard Cite

Abstract-Irregular streaks are important clues for Melanoma (a potentially fatal form of skin cancer) diagnosis using dermoscopy images. This paper extends our previous algorithm to identify the absence or presence of streaks in a skin lesions, by further analyzing the appearance of detected streak lines, and performing a three-way classi cation for streaks, Absent, Regular, and Irregular, in a pigmented skin lesion. In addition, the directional pattern of detected lines is analyzed to extract their orientation features in order to detect the underlying pattern. The method uses a graphical representation to model the geometric pattern of valid streaks and the distribution and coverage of the structure. Using these proposed features of the valid streaks along with the color and texture features of the entire lesion, an accuracy of 76.1% and weighted average area under ROC curve (AUC) of 85% is achieved for classifying dermoscopy images into streaks Absent, Regular, or Irregular on 945 images compiled from atlases and the internet without any exclusion criteria. This challenging dataset is the largest validation dataset for streaks detection and classi cation published to date. The data set has also been applied to the two-class sub-problems of Absent/Present classi cation (accuracy of 78.3% with AUC of 83.2%) and to Regular/Irregular classi cation (accuracy 83.6% with AUC of 88.9%). When the method was tested on a cleaned subset of 300 images randomly selected from the 945 images, the AUC increased to 91.8%, 93.2% and 90.9% for the Absent/Regular/Irregular, Absent/Present, and Regular/Irregular problems, respectively.

show abstract

Section: Methods For Three-class Classification For Streaks Detecmentioning

confidence: 98%

Section: Previous Workmentioning

confidence: 99%

Section: Previous Workmentioning

confidence: 99%

Section: A Identifying Valid Streak Lines From Candidate Streaksmentioning

confidence: 99%

mentioning

confidence: 93%

See 3 more Smart Citations

Detection and Analysis of Irregular Streaks in Dermoscopic Images of Skin Lesions

Sadeghi

Lee

McLean

et al. 2013

IEEE Trans. Med. Imaging

112

View full text Add to dashboard Cite

show abstract

Automated Detection of Streaks in Dermoscopy Images

Delibasis

Kottari

2015

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Part 1: Image Processing and AnalysisInternational audienceIn this paper we present a novel algorithm for the detection of dark linear structures, which appear in digital dermoscopy images of skin lesions and they are called as streaks in relevant literature. The proposed algorithm is capable of detecting such linear structures using local image curvature information obtained by the Hessian matrix. A linear structure is characterized as streak, based on its geometric characteristics. The streak detection algorithm is applied to a number of dermoscopy images containing malignant and non-malignant skin lesions. In this work we propose also a new class of streak based image features and we investigate their value in image classification according to malignancy

show abstract